hide You are viewing an archived web page, collected at the request of Ethereum Foundation using Archive-It. This page was captured on 04:24:53 Jun 24, 2021 , and is part of the Community collection. The information on this web page may be out of date. See All versions of this archived page. Loading media information

Ethereum user interface - rapid iteration and crowd testing - great design as key for success

I had yesterday a great email discussion with @asva. He suggested to post a part of our discussion on the forum.

As a member of the Apple developer community for me is the UI the starting point for every great coding project. Thus I am exited to learn more about the present stage of the UI developing and the persons / team involved. Who are the contact persons?

An Apple strategy is to have the UI done before touching any code. Apple is using 2/3 of their project time for UI design. They first start coding after the UI is done - avoids spaghetti coding.

Looks like, that still a lot of exiting UI work is ahead, although some nice work has been done. In my view, the goal is an self explaining UI which do not need any user manual. is focused on the main feature and easy and intuitive to use. I think this is possible.

I do not know, if you have access (or maybe you know them already) to Apples dev videos WWDC 2014 https://developer.apple.com/videos/wwdc/2014/
"Prototyping - fake it until you make it"
provides an awesome recipe for rapid UI development. Is Stephan doing the UI testing with the crowd?
Another great talk
"Designing Intuitive User Experience"

The design is key for the success of the project. An easy to use and main feature focused design will decide, if we or somebody else who fork the project will succeed among the users. I am very interested in the design question, as I am very interested in design and Art in general. My wife is a professional painter.

What about an unified UI for all implementations? The team is still quite small, and this it might help to reduce redundancy in the work. What about to focus on one UI for the desktop / laptop version (and one for genuine mobile and one for the mobile web app)?

Looking forward to learn more about the UI project. Is anybody doing UI prototyping and rapid iteration based on user feed back?


  • Gerd_HGGerd_HG Member Posts: 42
    Who has interest to help with the user interface for: iEthereum: An easy to use Ethereum iPhone and iWatch implementation for the general public. see GitHub https://github.com/iEthereum
  • tgerringtgerring Member Posts: 28 ✭✭
    Hey Gerd_HG,

    UI/UX is something the entire team is passionate about and it probably shows when you see many of the developers working on Macbooks :)
    Gerd_HG said:

    In my view, the goal is an self explaining UI which do not need any user manual. is focused on the main feature and easy and intuitive to use. I think this is possible.

    While I agree in theory, reality often proves to be much different and requires that UX shifts be introduced gradually to train people how to interact with the new system. As that base knowledge grows, we can build more advanced complexity because most users will already "get it". Consider the evolution of iOS itself: The large touchscreen with siloed apps what groundbreaking in many ways, but also caused quite a bit of confusion for non-technical people. It would have been foolish to introduce the App Store and all the UI/UX inconsistencies it breeds. But here we are several years later and even my grandma knows how to download apps.

    Furthermore, I'm seeing what I'll call "philosophical" differences when it comes to discussing UI internally. Some people want a very technical interface for maximum information display. Others want an extremely traditional flat 2D interface. Yet others want to push the envelope to a more 3D/HUD style. So how do we combine these all into one that everyone can agree on?

    In the end, I don't think we can... and this makes sense in the context of our ideal decentralized universe. Considering that we have 3-4 "main" clients (C++, Go, Python, Java), I believe there will be a lot of room for innovation in a variety of UX styles. And ideally none of this will matter since the clients can serve as a dumb JSON-RPC interface, allowing even more third-party interfaces to be developed.

    As we spin up development with new resources in hand, I hope to see much more rapid iteration of the interface to find one that at least *I'm* comfortable with. And although they may appear to quite broken, considering they've evolved to this point during a period in which brand new technology is being created and demoed to the world. The fact that they have GUI at all amazes me :)

    Do you have any specific ideas/prototypes/mockups you'd like to see incorporated?
  • vaXvaX Austin, TXMember Posts: 78 ✭✭✭
    tgerring said:

    reality often proves to be much different and requires that UX shifts be introduced gradually to train people how to interact with the new system.

    "philosophical" differences . . . makes sense in the context of our ideal decentralized universe.
    ideally none of this will matter since the clients can serve as a dumb JSON-RPC interface, allowing even more third-party interfaces to be developed.
    The fact that they have GUI at all amazes me :)
    Ẏ☰ϟ! Ẏ☰ϟ! Ẏ☰ϟ! Ẏ☰ϟ!
  • Gerd_HGGerd_HG Member Posts: 42
    @tgerring I think one needs first to identify the main feature and totally focus on it. I can agree with you, that over time one can add new fuctionality. At the first step focus on one feature and make this really great.

    Who decides which UI? - In my view the user. It is very important to make prototypes and constantly test them with the user. The talk I cited above provides a good recipe.

    I am working on an UI - hoverer only for the iPhone and iWatch - light client. (as company CytoComp is developing a biosensor / effector which works with this hardware)

    I will be very happy to share prototypes. If you follow the iEthereum GitHub https://github.com/iEthereum, you will be early informed.

    If you are very interested, I am always open for a G+ hangout or Skype.
  • Gerd_HGGerd_HG Member Posts: 42
    Ẏ☰ϟ! Ẏ☰ϟ! Ẏ☰ϟ! Ẏ☰ϟ! = :-) :-) :-) :-) :-) :-) :-) :-) @vaX
  • Gerd_HGGerd_HG Member Posts: 42
    @vaX‌ and @tgerring‌ One idea I have in respect to UI is to provide the user with standard contracts - for the non coder.... One could e.g. think about 10 exiting one. What about a competition in the community to provide this. To make it attractive one could give e.g. 20BTC and 5000 Ether to the top 10 winning contributions. The crowd funding / "IPO" of ethereum collected so much unexpected money that there should be financing for this.
    I think the etherum app should wow at launch time. It will, if it is easy to use. People will quick desinstall the app, if we fail. We wish to get an as broad audience as possible.
  • BezpolBezpol Member Posts: 26
    Ger_HG - you are right. Client makes decission. So it is not easy to make UI which will be good for endusers and more important is help understand for client how it works and for what it is ... Clients like new ideas but if they don't see profits in short times they kill the idea. I think that we need find one good place for use ethereum and make big work for perfect work and don't say that we can use ethereum everywhere. Everywhere is nowhere.
  • Gerd_HGGerd_HG Member Posts: 42
    edited September 2014
    @Bezpol‌ agree. I think we have basically 2 target groups: 1) the Daap developer 2) the contract user. Thus we have two application definitions, thus maybe 2 basic types of apps. We need a developer app and an app for the general audience. This means also two user interfaces. Initially one could e.g. build one app for developers used on a e.g. Laptop and one mobile light version for the general audience. In any case, both apps will need a very focused UI, to begin with one main feature / focus. We need to simplify. This might also answer some criticism ethereum got see http://www.cryptocoinsnews.com/gavin-andresen-outlines-bitcoin-2-0-without-ethereum/ Gavin Andresen "went as far as to predict that ethereum would either radically reduce their scope or forever play whack-a-mole with security" This will also help the general development process of the project, avoids spagetti coding, redundancy.... The app (also the developer one) has to work right out the box wo any manual or long instructions. We need also start to test UI`s with users and rapid iterate due to the feedback.
    I think this procedure is key for a success of ethereum. Ethereum builds on a P2P network. If we wish to have a decentralized network we need as many as possible users, otherwise the system gets centralized towards a few developers. A wide audience demands ease in use, thus an intuitive UI.

    What do you think?

    What about a competition where we can send in UI mockups. What about a bounty - there should be plenty of money from the crowd funding "IPO"
  • Gerd_HGGerd_HG Member Posts: 42
    edited September 2014
    PS https://github.com/voisine/breadwallet a BTC light wallet, which might be an example how a focused and easy to use UI can look
  • FrodonomicsFrodonomics Member Posts: 14
    edited September 2014
    As an end user I'd want:

    - Easy to use apps with few options.

    - Click on an 'advanced' tab for deeper interaction. The 'spitfish'/blockfish compressor for audacity is my favourite GUI.

    - Big, slightly cute buttons, choice of skins. (You might not care about colors, but my gf does)

    - Etherlite, etherbasic, etherpro. Users understand these phases. Ether v11.55 betafnce.exe is elitist and exclusive. I understand you're not there yet, Im just looking forward to Christmas.

    - For complicated operations, a helper text when you hover the mouse is invaluable. An AVI compressor called handbrake did this brilliantly.

    - Standard contracts and safety would help. The first month I use ethereum Im going to be thinking "Mt Gox, Mt Gox, Mt Gox"- it doesn't matter if its right or wrong. Alt coin paranoia is losing you 80-90 percent of Asians. That is, if you care about the Asian markets.

    - For me, 3-5 buttons is perfect. I just want to buy/sell ebooks and buy mp3s.

    - I don't think you need a one size fits all app, a suite of individual apps is much more attractive. Again, see Spitfish.

    - You might all have Macs, but most of us run android. Theres an android problem where cheaper phones wont run bigger apps, regardless of how much space you clear... That bug is well known. So, smaller is the way forward.

    - "Ethereum launch plagued by bugs and unwieldy GUI" Your competitors are waiting to write that. (!)

    - For Asia, use cute Ninjas, except in China.

    - OK. Im new here, don't know if I'm way off, but that's what I think.

  • Gerd_HGGerd_HG Member Posts: 42
    @Frodonomics‌ I totally agree with your view, exactly as I see this too.

    We need platform specific solutions. In respect to this we might wish to discuss a mobile web app with responsive design. The disadvantage of such a setting might be the requirement of a server, which we wish to avoid in our project - but maybe as an additional option, although it looks like that people prefer genuine apps on their mobile devices. Thus, I have so far no clear opinion on a mobile web app - looking forward on comments on this.

    It looks like, that we will end up with several ethereum versions. The question is on which ones we should focus on first, as we obviously can not do all at the same time. I think we might need an app for developer - obviously to generate awesome Daap`s people can use. Moreover, I see on Google analytics of my web apps that about 70% of my users use mobile. Thus a genuine mobile app (light version of ethereum) (for the general audience) for iOS and android might be another priority. What do you think?

    Safety is also an important point. As you mention, simplification can help to make the app more safe.

    I think simplification, ease, focus and great UI design looks like a common theme for our design goals.
  • Gerd_HGGerd_HG Member Posts: 42
    PS mobile web apps might also have a security flaw - same as web apps for BTC - target for hackers of all kind
  • FrodonomicsFrodonomics Member Posts: 14
    edited September 2014
    Gerd, good thinking. If you bullet point your q's it helps.

    I need a place with infinite edits so I can tweak my posts. I also need a community that understands Asia is a vibrant market for the future.

    When you say app, what are you thinking? I could reel off a whole list of one use apps right now. 5 that are a priority, 15 that are useful.

    So what's your priority?

    I'm mainly thinking along the lines of super simple decentralized kickstarter, twitter, ebay type programs.

    I'm really looking for a community of coders that want to solve real world problems one by one. Where can I hang out?

    Is it hard for you to port from iOS to android?

    The only comment I have for safety is A) Do your best. Bb) We (consumers) are going to have to learn self responsibility and caveat emptor again. C) Keep first release max account limits low. 10 bucks max storage. 3 bucks per transaction. Better to irritate people with small transaction limits than deal with a pitchfork mob who lost their savings.

    I understand that decentralized cryto 'banking' means another Mt Gox can't happen again. But does my neighbour?

    If you tell me what apps most interest you we can move forward.

    UI is important but I wouldn't obsess over the sleek look, especially if it costs you bytes. A simple tweak of a generally available open source skin is fine. I hate fancy graphics that slow down my comp and phone...

    OK. More random thoughts. Good to see a programmer that respects the non tech user...

  • FrodonomicsFrodonomics Member Posts: 14
    Last thing, maybe nonfinance apps are cool too: Meetup, communication, sharing music rapidly, crowdtranslation....
  • Gerd_HGGerd_HG Member Posts: 42
    @Frodonomics‌ Thanks for your great reply. Great suggestions for potential apps. When I talk above about the Ethereum app - I have the ethereum platform per se in mind.

    I have a personal interest to make a light version for the iPhone and the Apple watch see
    I just started this one. I will soon add a user interface mockup. Thus if this interests you feel free to follow https://github.com/iEthereum/iPhone----Apple-Watch---ethereum
    This is my first focus.

    In respect of Dapps I am interested and started to explore a bit in a DNA analysis Daap https://github.com/CytoComp/DNA-analysis---ethereum---Dapp and to build a decentralized company https://github.com/CytoComp/decentralized-company-ethereum-Dapp I shared some thoughts in the Project part of this forum.

    I am an Apple developer and objective-c and lately swift are my speciality. In respect to android, I would need to team up with other. If we talk about the android version of the ethereum platform, I think this should have very high priority for the community. Somebody needs to work on this.

    Asian market - agree that this is very important. Great apps are regionalized (voluntary - the customer should be able to make the settings; not enforced regionalization, you should be able e.g. to use a US version everywhere, if you wish to, or an asian version if you are e.g. in the US).

    In my view is coding about solving problems for people. Many people become quite good in writing code after some years. What makes a world class coder is to focus first on the identification of a large problem. Than to find the solution. Than make a great user interface (all these steps needs constant feedback from the user and rapid iteration due to the user feedback). Thus the first steps are designing abstraction. First when this is done one should start to code. These first steps should use 2/3 rd of your total project time. 1/3 goes to finding the technical solutions / coding and the last step is debugging. BTW this is the way they teach you computer Science at Apple or Stanford University - and these people should know....

    I think you have great Dapp ideas. The next step might be to figure out on which to focus. Which one do you have the biggest passion for? Which one solve the biggest problem, the largest number of people possible have?
  • FrodonomicsFrodonomics Member Posts: 14
    edited September 2014
    I follow and agree with pretty much all you say.

    I realized I was jumping ahead and brainstorming Dapps before the platform has been coded.

    The fact is I need infinite edits to work properly, otherwise I end up repeating myself. I want to leave clear ideas for coders to follow, so I need to be able to modify posts.

    I'll go and start a thread called "Brainstorming blockchain apps for the future". I'll do it on bittalk unless there's anywhere better. Then I can get my ideas together and ppl can read, if they want, in their own time.

    Maybe Dapp ideas won't be relavent til winter, as you guys are on step 1.

    Im interested in solving issues like where to buy low cost water filters, solar dehydrating food tutorials, a "craigslist for basic needs technology" moving money without tariffs, decentralized governance, meeting other changemakers, bringing crypto to non tech users, rapid file sharing etc.

    Education and communication is key.

    I'm interested in all of those potential apps. I think ethereum can cover half of those. The key is getting the ideas out there so ppl can code anything that interests them.

    If an app is small and intuitive, I can translate it to Indonesian, Spanish and Mandarin. That's 2 billion ppl in up and coming markets. I have a fair idea how to localise.

    I'll go check those links and start a brainstorming thread. I like the ideas here but I don't want to derail any work-in-progress by talking about Dapps when the platform isn't there yet!

    Post edited by Frodonomics on
  • Gerd_HGGerd_HG Member Posts: 42
    @Frodonomics I really love to learn about problems which need to be solved. Please post the link to thread on bittalk here. I think we need to dart with the Daaps right now. As I described in the design process which e.g. is done in a similar way by Apple you will need in a 3 month project 2 month for the design and 1 month for coding. The design process is about first to identify the problem - in economical terms a 1 billion $ total market size problem, than find the solution, make sketches of the solution, and make design prototypes which are rapid iterated. If you can get access to the Apple WWDC video "Prototyping - fake it until you make it" https://developer.apple.com/videos/wwdc/2014/#223-video you can get a great introduction to the process. BTW these first steps can be done wo any coding knowledge. If you google there are several prototyping apps out for other environments than Apple, in case you need. As you have so many great ideas, why not work on them.......
  • JoshuaJBerkJoshuaJBerk Brooklyn, NYMember Posts: 12
    Perhaps worth mentioning that, as a primarily developer product, the main UI is the API, not the GUI. Although I'm sure the team seeks to make it as user-friendly as possible to enable any level of sophistication in Đapps, the important thing to remember is that a good UI is as little as it must be but no more. Their best hope of that is what the DEV team is doing already -- reducing complexity of the underlying code. Perhaps I'm mistaken, but I don't anticipate someone having an Ethereum wallet in the same way as one might with Bitcoin but rather an AppleCoin and AppleWallet with ETH as the core of the application's plumbing.

    @Gerd_HG‌ -- IMO your time would be best spent on the projects page, helping the Đapp creators polish their ideas. :)

    But, on a very basic level, I'm 100% onboard with a UI-first approach -- pays compounding dividends over time. It's just *which* UI we're talking about!
Sign In or Register to comment.